Tasks involving the comprehension and imitation of sentences have reliably distinguished children and adults with specific language impairment (SLI) from their typical language peers (Conti-Ramsden, Botting, &Faragher, 2001;Tomblin, Freese, &Records, 1992) and are widely used in clinical evaluations for language disorders (Hammill, Brown, Larsen, &Wiederholt, 1994;Semel, Wiig, &Secord, 2003). Despite compelling findings on the effectiveness of sentence imitation as a diagnostic tool, little is known about the factors that underlie performance differences on this task for individuals with SLI. The aim of this research is to specify the role of three factors hypothesized to play a role in sentence imitation performance, leading to a better understanding of why the task effectively diagnoses SLI, and how SLI affects young adults. The research program will involve the comparison of a group of young adults with SLI to young adults with typical language. Experimental tasks will measure working memory and language processing speed. In a sentence comprehension task, the degree of processing difficulty for phrases that are either arguments (such as "the ball" in "Bob hit the ball") or adjuncts (such as "over the fence" in "Bob hit the ball over the fence") will be assessed. Participants will also imitate sentences that vary in memory demands and the use of adjunct, or elaborating, constituents. Analyses of the results will provide evidence on whether adults with SLI have relatively intact lexical systems and relatively impaired syntactic processing systems, as predicted by a recent theory on the causes of SLI (Ullman &Pierpont, 2005). The study will also indicate if working memory, language processing speed, and adjunct processing difficulties each contribute to sentence imitation performance, and the extent to which each factor contributes to performance in combination with the others. PUBLIC HEALTH RELEVANCE: Specific language impairment (SLI) affects more than 7% of children in kindergarten and the condition persists into adulthood in over 70% of cases, with significant consequences for the educational, vocational and social well being of affected individuals. Sentence imitation performance is one of the most effective and widely used means of identifying individuals with SLI, but little is known about why it is so effective. This project aims to identify specific factors that may explain the poor sentence imitation performance of adults with SLI, indicating how the task can be refined for better diagnostic accuracy and clarifying what poor sentence imitation reveals about the nature of SLI.